Another Perspective on Vocabulary Richness

This article examines the usefulness ofvocabulary richness for authorship attributionand tests the assumption that appropriatemeasures of vocabulary richness can capture anauthor's distinctive style or identity. Afterbriefly discussing perceived and actualvocabulary richness, I show that doubling andcombining texts affects some measures incomputationally predictable but conceptuallysurprising ways. I discuss some theoretical andempirical problems with some measures anddevelop simple methods to test how wellvocabulary richness distinguishes texts bydifferent authors. These methods show thatvocabulary richness is ineffective for largegroups of texts because of the extremevariability within and among them. I concludethat vocabulary richness is of marginal valuein stylistic and authorship studies because thebasic assumption that it constitutes awordprint for authors is false.     

综合应用地震属性、测井数据反演储层参数

本文研究了样点地震属性的特征提取，利用BP神经网络建立测井声波时差和地震属性的非线性映射关系。采用距离加权平均对研究区所有井的反演结果进行外推。为了提高分辨率和计算结果的准确性，详细分析、研究了适合于地震信号分解和重构的小波，以及测井声波时差和地震波阻抗与层界面的关系。最后，对反演的声波时差资料进行界面影响校正，进而计算出较符合实际情况的孔隙率参数。经已知井和新钻井验证，计算结果和实际情况相一致，效果令人满意。     

Specific and general attributional patterns of Holocaust survivors.

Members of Holocaust survivor groups and participants at survivor gatherings were asked to fill out questionnaire batteries, which included two measures of attribution. One asked their views of the factors that had led to survival during the Holocaust; the other was the Attributional Style Questionnaire (ASQ), a standard measure of attributional styles. The results were compared with the responses of age-matched Jewish respondents who had spent the period of the Holocaust safe from Nazi persecution (mostly in North America). Concerning Holocaust survival, survivors mentioned significantly more factors, and 91% of survivors but only 51% of controls cited external factors (e.g., luck, help from others), with the reverse pattern (71% vs. 34%) for internal factors such as psychological strength and determination. There were no significant intergroup differences on the Attributional Style Questionnaire as a function of sex, age, or survivor versus comparison group membership. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Incorporating If...Then...Personality Signatures in Person Perception: Beyond the Person-Situation Dichotomy.

Three studies investigated conditions in which perceivers view dispositions and situations as interactive, rather than independent, causal forces when making judgments about another's personality. Study 1 showed that perceivers associated 5 common trait terms (e.g., friendly and shy) with characteristic if...then... (if situation a, then the person does x, but if situation b, then the person does y) personality signatures. Study 2 demonstrated that perceivers used information about a target's stable if...then... signature to infer the target's motives and traits; dispositional judgments were mediated by inferences about the target's motivations. Study 3 tested whether perceivers draw on if...then... signatures when making judgments about Big Five trait dimensions. Together, the findings indicate that perceivers take account of person-situation interactions (reflected in if...then... signatures) in everyday explanations of social behavior and personality dispositions. Boundary conditions are also discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Comparing techniques for authorship attribution of source code

Attributing authorship of documents with unknown creators has been studied extensively for natural language text such as essays and literature, but less so for non‐natural languages such as computer source code. Previous attempts at attributing authorship of source code can be categorised by two attributes: the software features used for the classification, either strings of n tokens/bytes (n‐grams) or software metrics; and the classification technique that exploits those features, either information retrieval ranking or machine learning. The results of existing studies, however, are not directly comparable as all use different test beds and evaluation methodologies, making it difficult to assess which approach is superior. This paper summarises all previous techniques to source code authorship attribution, implements feature sets that are motivated by the literature, and applies information retrieval ranking methods or machine classifiers for each approach. Importantly, all approaches are tested on identical collections from varying programming languages and author types. Our conclusions are as follows: (i) ranking and machine classifier approaches are around 90% and 85% accurate, respectively, for a one‐in‐10 classification problem; (ii) the byte‐level n‐gram approach is best used with different parameters to those previously published; (iii) neural networks and support vector machines were found to be the most accurate machine classifiers of the eight evaluated; (iv) use of n‐gram features in combination with machine classifiers shows promise, but there are scalability problems that still must be overcome; and (v) approaches based on information retrieval techniques are currently more accurate than approaches based on machine learning. Copyright © 2012 John Wiley & Sons, Ltd.     

基于测井资料和地震属性分析地震相分布

地震相是特定沉积相或地质体在地震剖面图上所反映的主要特征的总和,是沉积和地震之间联系的枢纽。利用不同相带的井在地震剖面上的标定结果和声波时差统计,在常规地震剖面和瞬时相位剖面初步得到相带划分图。根据地震反射层的外形结构、频率、振幅及连续性等特征,建立三种沉积相对应的地震相模式,分析沿目标地层顶界地震属性切片,筛选对相带边界表征清晰的属性,确定出地震相的分布区域,这对以后的储层预测和生物礁的发现提供技术支持。     

Longitudinal test of ePortfolio continuous use: an empirical study on the change of students’ beliefs

The electronic portfolio (ePortfolio) is learner-centred, and its effectiveness depends on the learners’ long-term use. Thus, it is proper to conduct the study from the learners’ perspectives. Currently, most research on the use of the ePortfolio is in the form of cross-sectional studies, and it is difficult to find the reasons for the changes in students’ beliefs with regard to the long-term use of the ePortfolio. Based on the longitudinal test, this study explained the students’ continuous use of ePortfolio and the changes in their beliefs. With 122 samplers in the adoption–continuous stage (t ₁–t ₂) and 117 samplers in the continuous stage (t ₂–t ₃), this study demonstrates that in continuous use stage (t ₂–t ₃), perceived ease of use still influences the users’ perceived usefulness (PU) and attitude towards the ePortfolio. Attribution significantly moderates the users’ beliefs from the adoption stage (t ₁) to the continuous use stage (t ₂). However, the moderating effect of attribution in the continuous use stage (t ₂–t ₃) is insignificant; satisfaction and attribution are the key factors driving the users’ continuous intention towards the ePortfolio rather than PU and attitude. Meanwhile, satisfaction and attribution will change because of the users’ expectation disconfirmation, and it will influence the users to continue using ePortfolio.     

冰冻圈水文过程对黑河上游径流的影响分析

黑河上游处于高寒山区,开展冰冻圈水文过程对径流的影响研究,对于判断该地区径流变化趋势及其可持续性具有重要意义。本文应用基流分割、逐步多元回归等方法,分析了1960—2013年黑河上游出山径流量的变化及其原因,重点估计了融雪、冰川融化对径流的贡献,探讨了土壤冻融过程对径流变化的可能影响。分析结果表明,黑河上游山区河川径流在近54年间呈现上升趋势,其中以基流部分的上升为主。降水与气温变化对河川径流上升均有较大贡献。导致春季径流增加的主要因素为温度、降水次之;导致秋季径流增加的主要因素为降水、温度次之。夏季黑河上游东支的雨强有显著增大趋势(日雨量增幅为0.9 mm/10 a),对于河川径流贡献的比例约为15%。黑河山区的逐年降雪量变化不显著,冰川融化对径流增加的贡献小于10%。由此推断气温上升导致基流增加的主要原因是:气温升高导致高寒山区冻土活动层增厚,增加了土壤蓄水容量,从而导致降雨下渗量增加和基流量增大。由于黑河上游冻土分布广泛,未来气温持续上升的情况下,这种产流机制变化导致的基流增加具有可持续性。     

基于BP网络的洪灾风险评价算法

为了解决洪灾风险评价问题,文中在分析传统洪灾风险评价算法不足的基础上,结合BP网络特点,提出了基于BP网络的洪灾风险评价算法（Flood Risk Evaluation algorithm on BPNet,FRE—BPN）,同时为了解决BP算法对复杂高维样本数据建模效率差、精度低的缺陷,利用粗糙集对BP网络进行约简。仿真实验表明,粗糙集约简使得BP网络大大降低了洪灾风险评价模型建立的复杂度,同时FRE-BPN算法对洪灾数据集进行风险评价时的平均耗时大大下降。     

IoT设备程序同源性智能检测技术综述

IoT设备与各行业的深度融合方兴日盛,使得IoT程序快速开发的需求不断增长。开发者习惯于集成第三方库或常用代码。不幸的是,若这些代码中隐藏着潜在的漏洞,那他们也会被扩散到不同的程序中,为其大规模扩散创造了条件。这也是造成近年来IoT设备群体性安全事件频发的重要原因之一。为了降低危害,发现具备相似漏洞的程序并进行相关的处置是一个有效的方法。同源性分析作为挖掘程序间关联关系的重要手段之一,可高效地实现程序漏洞的智能溯源取证。结合机器学习和深度学习技术,它表现出解决大规模程序安全性检测的巨大潜力。然而,IoT设备的软硬件特点仍使得该技术的使用面临挑战。当前已有诸多方案在IoT设备程序的同源性智能检测方面取得了进展。因此,本文将系统性回顾近年来相关技术研究的成果,将他们分为相似性分析和创作者归属技术。首先,我们介绍了两种方式的数据来源。接着,检测过程中涉及的特征选择、特征表示以及相对应的检测方法也被依次介绍。进一步的,本文不仅比较和总结了方案的特点和局限性,还对他们在不同类型IoT设备程序的适配性进行了对比分析。最后,文章针对IoT程序分析提出了一些研究建议。作者希望本综述可为研究者阐明这些工作的核心技术点,并为他们在IoT设备上的进一步应用提供启发。